STICKLEBACKS. 85 



Poiss. to be insufficient to mark it with certainty as a species distinct 

 from the full-armed Gasterosteiis of Europe or America (p. 500). 



Dr. Parnell, in his Fishes of the Frith of Forth (p. 34), after stat- 

 ing that he agrees M'ith Cuvier and Yarrell in considering the G. trachurus 

 as " a constant and well-marked species," observes that the " square tail " 

 does not exist in the other sticklebacks. According to my observation, 

 it is certainly less developed in them, and generally (but not invariably) 

 corresponds with the protecting side-plates, presenting a greater or less 

 development accordingly as the armature of the body is of a heavier or 

 lighter cast. IJr. Parnell further remarks, as corroborative of G. trachurus 

 being a distinct species, that he has " examined carefully several hundred 

 from half an inch to two inches and a half in length, and in all the speci- 

 mens the lateral plates were constant." In particular localities I have met 

 Avith the same result on examining specimens of all sizes of G. trachurus 

 and of the other varieties also,* but in some places again the different 

 varieties are found together and of every size.f Mr. Yarrell has so 

 noticed three of them in the Thames at Woolwich ; and in Pathlin, as 

 before mentioned, they occur together — in the former locality in brackish, 

 in the latter in fresh, water. 



G. semiarmatus, Cuv. and Yal., t. iv. p. 493, appears to be the rarest of 

 the 3-spined sticklebacks in Ireland. I possess specimens from the is- 

 land of Rathlin, as before mentioned, and from Wolfhill, in the neigh- 

 bourhood of Belfast. One example only occurred in the latter locality, 

 Avhereit was taken in 1832 with a number of G. hrachy centres, t\\e stickle- 

 back of that district ; it is, indeed, this variety in every respect, except in 

 having the lateral plates extending along the sides so far as in G. senii- 

 armatus ; the other characters assigned to this supposed species in the 

 Hist, des Poiss. are very variable. From the half-armed species I turn to 

 the 



G. leiurus, Cuv. and Yal.,t. iv. p. 487, in which the lateral plates do not 

 extend beyond the pectorial region. In every respect but this it is con- 

 sidered in the Hist, des Poiss. so similar to G. trachurus, that the one 

 description is given as equally applicable to both. The'G. leiurus Avould 

 seem to be the most conwuow freshivater stickleback in Ireland. J 



The localities whence specimens of this fish are now before me, are — the 

 island of Kathlin ;— the neighbourhood of Belfast (W. T.) ; — Kiver Bann 

 at Toome (W. T.) ; Portaferry and Newcastle, County Down (W. T.) ; — 

 Lough Melvin, County Fermanagh (W. T.) ; — neighbourhood of Dublin 

 (Dr. Ball) : — Glendalough, County \Yicklow (Mr. G. C. Hyndman) : — Port- 

 arlington. Queen's County (Ptev. B.J.Clarke) ; — some of the examples from 

 this locality are very handsomely marked, being along the back of a rich 



* The partial exception to this is in G. brachycentrus, in which the dorsal 

 spines are comparatively longer in young than in adult individuals, and hence tlie 

 young in this respect accord with G. leiurus. I here speak of localities hi which 

 all the lull-grown fish are G. brachycentrxis. 



t From the many small examples of all the varieties about nine lines in 

 length that have come under my observation, I should think the number of 

 lateral plates they are to possess through life is then as decided as the number of 

 fin-rays ; i. e. jirovided they would have remained in the locality wlieiice they 

 were taken. Whether such a change of habitation, as from fresh water to the 

 sea, would cause the smooth-sided at any age to put on the lateral armour may 

 remain a question. 



X From drains which are occasionally replenished by the tide 1 have also 

 taken it. 



